Relevant to our research interests in animal models of neuropsychiatric diseases, our Laboratory of Behavioral Neuroscience investigated the behavioral phenotypes of transgenic and knockout mice with mutations in genes expressed in brain pathways involved in neuropsychiatric disorders. Collaborations with molecular genetics laboratories enhanced our access to targeted gene mutations of particular interest. Over the past few years, we have developed and refined a multi-tiered strategy for mouse behavioral phenotyping. We have also developed new behavioral tasks for mice and adapted rat behavioral tasks for mice. [unreadable] [unreadable] Dr. Crawley began a new initiative in 2002 to develop mouse behavioral tasks relevant to the symptoms of autism. Ultimately, these behavioral tasks will be used to identify genes that are uniquely expressed in strains of mice with social deficits. Approximately half of the autism-related experiments were conducted in collaboration with investigators at the University of North Carolina Chapel Hill (UNC), particularly Dr. Joseph Piven, Director of the Autism STAART Center. Mouse behavioral genetics experiments were designed to discover genes related to the social deficits that represent the fundamental symptoms of autism. Dr. Crawley and Dr. Sheryl Moy at UNC developed an automated three chambered apparatus that measures sociability and preference for social novelty in mice. George Dold and coworkers in the NIMH Research Services Branch built the prototype apparatus, based partly on previous methods in the literature. This year, Drs. Crawley and Moy and technicians in the UNC Neurodevelopmental Disorders Research Center completed testing of 20 inbred strains on the social tasks. Robust social interaction with a stranger mouse, as compared to interaction with a non-social novel object, were detected for most inbred strains, including C57BL/6J, DBA/2J, FVB/NJ, C3H/HeJ, AKR/J, PL/J, and SWR/J. Low social approach was detected in A/J, Balb/cJ, BTBR T+tf/J, 129SvImJ, and C58/J. Control tasks to measure general health, neurological reflexes, motor functions, sensory abilities, and anxiety-like behaviors indicated compromised procedural abilities in most of the strains with low social approach. BTBR T+Tf/J (BTBR) demonstrated a highly specific sociability deficit, with normal scores on all other procedural measures, including the elevated plus maze anxiety-related task and open field exploration.[unreadable] [unreadable] To further investigate the apparent social deficit in BTBR, Dr. Hewlet McFarlane, LBN guest researcher, tested BTBR and C57BL/6J (B6) control mice on a juvenile play task. Because autism is diagosed between ages 2 and 7 years old in humans, a comparable age of 21 days was chosen for mice. Using the Noldus Observer event scoring software, and choosing relevant parameters of juvenile social interaction from the literature, Dr. McFarlane scored multiple measures of social approach, following, sniffing, crawling over/under/past, and non-social behaviors during a 30 minute test session in the Noldus video chamber. Pairs of juvenile male BTBR displayed significantly less social approach, nose-to-nose sniffing, and social grooming than pairs of juvenile male B6. Further, our collaborator Dr. Valerie Bolivar discovered deficits in social transmission of food preference in BTBR. These findings of a highly selective social deficit in BTBR raise the possibility that this commercially available inbred strain may represent a mouse model of autism for wide translational applications.[unreadable] [unreadable] A second approach toward identifying the genes mediating the core symptoms of autism is to evaluate the behaviors of lines of mice with experimentally targeted mutations in genes relevant to autism. Comprehensive behavioral phenotyping was conducted on two lines in the past year. Dr. Crawley and Howard Hughes student intern Thomas Chen found normal social approach and social olfactory behaviors in oxytocin knockout mice generated by Dr. W. Scott Young, NIMH. Dr. Crawley and Dr. Moy and coworkers at the University of North Carolina found normal social approach in a second line of oxytocin knockout mice generated at Baylor School of Medicine and previously studied by Dr. Larry Young, Jim Winslow, Tom Insel, and co-workers at Emory University. Our negative findings in both lines of oxytocin null mutants on general social approach underscore the selective deficit in social memory previously reported by the Emory researchers for the oxytocin null mutants from Baylor. [unreadable] [unreadable] Vasopressin receptor knockout mice generated by Drs. Heather Caldwell and Scott Young, NIMH, were tested on the social task by Dr. McFarlane and Howard Hughes student Tommy Chen. Low social approach was detected in the first cohort. A second cohort, living on a reverse light cycle and tested during the dark phase of their circadian cycle, did not show a genotype difference on the social approach task. Dr. McFarlane conducted juvenile play analyses on the V1B mice and is presently analyzing the results. Dr. Maria Luisa Scattoni, guest researcher in LBN and an expert on mouse pup vocalizations, has measured vocalizations in V1B pups separated from their mothers at ages 7 to 12 days. No genotype differences were detected with either standard separation or on a more cognitive paradigm to evaluate maternal potentiation of pup ultrasonic vocalizations. The possibility that phase of the circadian cycle yields divergent genotype differences on social and communicative behaviors in young mice is being further investigated in our laboratory.[unreadable] [unreadable] Staff Scientist Joanna Hill, postbaccalaureate Maria Lim, and student volunteer Madeline Stone conducted the 3-chambered social behavior task and a range of developmental behavioral tasks on knockout mice deficient in vasoactive intestinal peptide (VIP), a developmental neuropeptide that has been implicated in autism. Results indicate social and developmental deficits in mice raised by mothers who are heterozygous for VIP. These findings support interpretations that prenatal intrauterine levels of VIP, and/or parental behaviors dependent on VIP, affect the subsequent behavioral development of the offspring.[unreadable] [unreadable] In collaboration with Dr. Andres Buonanno, NICHD, a colleague in the PNRC building, our LBN postdoctoral fellow Kathy Bailey tested mice with a mutation in the neuregulin-2 gene on prepulse inhibition of acoustic startle. Neuregulin-1 has been implicated in schizophrenia. First results show a small but significant deficit in prepulse inhibition in neuregulin-2 null mutants. This finding indicates that, like neuregulin-1, the neuregulin-2 gene may similarly be worth exploring in the etiology of schizophrenia.